Re: [Glossary] Deprecated, Deprecation, Deprecate / Obsolete

Le 29 juin 2004, à 09:52, Al Gilman a écrit :
> deprecated: future versions are planned not to be upward-compatible
> as regards this feature.

	Yes because it is replaced by a another method inside or outside the 
technology. itself.

> Since the 'aural' media type is not made unnecessary by *current*
> (that is to say equally or more mature) technology, but rather is
> only _slated to be made unnecessary by planned but as yet immature_
> technology, it could be argued that it is too soon to say in a PR or
> Recommendation text that the 'aural' media type designation is
> 'deprecated' because its function is still valid in terms of the best
> available current W3C Recommendation.

Be careful

* Deprecated is not obsolete.
	- Obsolete means not supported anymore
* Deprecated means should be implemented, though the scenarios have to 
be explicit
	See http://lists.w3.org/Archives/Public/www-qa-wg/2004Jun/0086

One issue with CSS that we have with CSS (at different levels), there 
are no ways to identify which versions of the stylesheet you are using 
and that is a problem for example in validation.
	Which version of CSS do you use?

	CSS 1.0
	CSS 2.0
	... CSS 2.1
	... CSS 3.O

Here "..." means future.

It's also difficult a user agent to know which version of CSS is 
supporting. CSS WG could say it's *less* (doesn't solve all) a problem 
in an inclusive model.

+-------------+
| +---------+ |
| | +-----+ | |
| | | 1.0 | | |
| | +-----+ | |
| |     2.0 | |
| +---------+ |
|        2.1  |
+-------------+

where all the previous features are part of the new specification. But 
as soon as you start to deprecate, you need to be able to identify the 
version. So you know which version, you are addressing and how your 
implementation must deal with it.

The issue is without clear understanding, mechanism to handle 
deprecated features, there is a danger, which is clearly increased by 
no versioning of the technology.


> Since continuity of formatting property notation is clearly a concern
> [7] and this question is correctly being raised here about a
> consistent theory of change control across W3C Specification
> version-sequences, perhaps we should link the two threads.

> [6] http://www.w3.org/TR/CSS21/aural.html#aural-media-group
> [7] http://www.w3.org/2001/tag/doc/formatting-properties




-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager
*** Be Strict To Be Cool ***